That's why the LedgerSMB project uses Template Toolkit with <?lsmb ?> tags (that and [% %] causes problems in LaTeX).
<html> <body> <?mypitag ... ?> </body> </html>
That's an SGML document. It doesn't matter where the PI tag is, or what is in it, it's an SGML document.
<?printstuff
'<html> <body> <p>fun stuff here</p> </body> </html>'
?>
That's still an SGML document.
<?php
# logic goes here
?>
That's an SGML document.
This is not splitting hairs. The reason is relatively subtle.... When you are doing SGML stuff an extra newline or a space in front of the opening tag really doesn't have any effect but if you do any output in binary form, it's a mess.
Let me give you an example.
Suppose I have a GUI application that prints some XML fragments to standard output. The most natural way to do this in PHP is to write an XML pre-processor which happens to make decisions based upon operations in the GUI. In other languages you would print to standard output explicitly and the GUI would be primary. In PHP the XML output would be primary and the GUI would be secondary.
In fact your entire GUI is sort of contained in XML fragments to start with.
I'm talking about the CLI SAPI, NOT the Apache SAPI.
Also on the PI tag issue, I would rather keep my PI tags limited to the view or template. it leads to a lot better clarity. and your object classes are no longer technically SGML fragments ;-) This has some nasty side effects if you have a leading space before your opening PI tag and you aren't outputting SGML....
Oh, as a side note, by 'escape from the web' I guess what I really mean is 'escape from SGML-world.' There's a lot of the web that's not SGML-driven and there's a lot of the non-web that is (XML, DocBook SGML, etc).
Bottom line, PHP is a decent tool for working with SGML but it is highly specialized for this. You run into weird things working in SGML mode but not in non-SGML output with things like leading spaces before opening PI tags if you move out of that. Since this is lexically part of the language, I just don't see it changing.
PHP started out as being a pre-processor for SGML documents. All PHP code is contained inside SGML program instruction tags. PHP has never changed this, so however it has developed it is still, in essence an SGML preprocessor because even today, all PHP code is contained in SGML PI tags.
This means that your class files, etc. all begin with opening a PI tag....
Now that doesn't mean there aren't some really cool niches for PHP. Using something like PL/PHP and the native XML support in PostgreSQL, you could do some pretty cool stuff. Similarly for lightweight web development there is simply not a better language.
But when you are building a framework where logic may have to be used in non-SGML environments, PHP is a very poor choice. This means when you choose PHP you are limiting yourself to web environments only. Better choices where there is any possibility of escape from the web include Perl, Python, and Ruby.